Three related experiments comprise this proposal. Their mutual aims to further our understanding of language and remembering in Alzheimer's Disease. The focus of the studies is on the analysis of discourse, and on the cognitive processes which may be viatal for remembering. The experiments are intended to 1) further characterize the nature of language and memory breakdown in AD; 2) clarify language/memory relationships in AD, and 3) provide some potentially useful insights into behavioral management of the disorder. Experiment 1 involves quantifying AD patients' performance of some simple tasks of daily living under varying conditions of verbal guidance from others, with the hope of providing management strategies that are hypothesis- driven, and of general utility to caretakers. Experiment 2 investigates the effects of differing discourse structures on AD patients' ability to remember verbal materials. Again, the hope is to provide some hypothesis-driven guidelines about how to frame language used to communicate with AD patients, to maximize their ability to remember. Experiment 3 is descriptive in nature, and will utilize techniques of narrative analysis to characterize verbal language in AD. This experiment will use language samples gathered for the most part in the first two experiments and will concentrate on describing coherence and cohesion in AD discourse.